A super-resolution algorithm using linear regression based on image self-similarity

Shen-Chuan Tai, Jiun Jie Huang, Peng Yu Chen

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The application of image super-resolution technologies in recent years has increased noticeably. The main purpose of image up-scaling is to obtain high-resolution images from low-resolution images, and these up-scaled images should keep satisfactory visual qualities and present natural textures. The most popular image up-scaling algorithms are based on interpolation methods in spatial domain. However, the up-scaled images may produce blurring artifacts. Therefore, using spatial sharpening filters is usually used to make blurred images sharp and clear. The quantity of image sharpening is the key to decide the visual qualities of up-scaled images. In this paper, a method based on self-similarity of images and using simple linear regression to build a reconstruction model for improving visual qualities of up-scaled images adaptively is proposed. The experimental results show that our algorithm provides better subjective visual qualities as well as the peak signal-to-noise ratio (PSNR).

Original languageEnglish
Title of host publicationProceedings - 2016 IEEE International Symposium on Computer, Consumer and Control, IS3C 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages275-278
Number of pages4
ISBN (Electronic)9781509030712
DOIs
Publication statusPublished - 2016 Aug 16
Event2016 IEEE International Symposium on Computer, Consumer and Control, IS3C 2016 - Xi'an, China
Duration: 2016 Jul 42016 Jul 6

Publication series

NameProceedings - 2016 IEEE International Symposium on Computer, Consumer and Control, IS3C 2016

Other

Other2016 IEEE International Symposium on Computer, Consumer and Control, IS3C 2016
CountryChina
CityXi'an
Period16-07-0416-07-06

Fingerprint

Super-resolution
Self-similarity
Image resolution
Linear regression
Signal to noise ratio
Interpolation
Textures
Upscaling
Simple Linear Regression
Interpolation Method
Texture
High Resolution
Filter

All Science Journal Classification (ASJC) codes

  • Signal Processing
  • Computer Networks and Communications
  • Computer Science Applications
  • Energy Engineering and Power Technology
  • Control and Systems Engineering
  • Control and Optimization

Cite this

Tai, S-C., Huang, J. J., & Chen, P. Y. (2016). A super-resolution algorithm using linear regression based on image self-similarity. In Proceedings - 2016 IEEE International Symposium on Computer, Consumer and Control, IS3C 2016 (pp. 275-278). [7545189] (Proceedings - 2016 IEEE International Symposium on Computer, Consumer and Control, IS3C 2016). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IS3C.2016.79
Tai, Shen-Chuan ; Huang, Jiun Jie ; Chen, Peng Yu. / A super-resolution algorithm using linear regression based on image self-similarity. Proceedings - 2016 IEEE International Symposium on Computer, Consumer and Control, IS3C 2016. Institute of Electrical and Electronics Engineers Inc., 2016. pp. 275-278 (Proceedings - 2016 IEEE International Symposium on Computer, Consumer and Control, IS3C 2016).
@inproceedings{9ac4da0b5c104ef6b756148fe816914d,
title = "A super-resolution algorithm using linear regression based on image self-similarity",
abstract = "The application of image super-resolution technologies in recent years has increased noticeably. The main purpose of image up-scaling is to obtain high-resolution images from low-resolution images, and these up-scaled images should keep satisfactory visual qualities and present natural textures. The most popular image up-scaling algorithms are based on interpolation methods in spatial domain. However, the up-scaled images may produce blurring artifacts. Therefore, using spatial sharpening filters is usually used to make blurred images sharp and clear. The quantity of image sharpening is the key to decide the visual qualities of up-scaled images. In this paper, a method based on self-similarity of images and using simple linear regression to build a reconstruction model for improving visual qualities of up-scaled images adaptively is proposed. The experimental results show that our algorithm provides better subjective visual qualities as well as the peak signal-to-noise ratio (PSNR).",
author = "Shen-Chuan Tai and Huang, {Jiun Jie} and Chen, {Peng Yu}",
year = "2016",
month = "8",
day = "16",
doi = "10.1109/IS3C.2016.79",
language = "English",
series = "Proceedings - 2016 IEEE International Symposium on Computer, Consumer and Control, IS3C 2016",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "275--278",
booktitle = "Proceedings - 2016 IEEE International Symposium on Computer, Consumer and Control, IS3C 2016",
address = "United States",

}

Tai, S-C, Huang, JJ & Chen, PY 2016, A super-resolution algorithm using linear regression based on image self-similarity. in Proceedings - 2016 IEEE International Symposium on Computer, Consumer and Control, IS3C 2016., 7545189, Proceedings - 2016 IEEE International Symposium on Computer, Consumer and Control, IS3C 2016, Institute of Electrical and Electronics Engineers Inc., pp. 275-278, 2016 IEEE International Symposium on Computer, Consumer and Control, IS3C 2016, Xi'an, China, 16-07-04. https://doi.org/10.1109/IS3C.2016.79

A super-resolution algorithm using linear regression based on image self-similarity. / Tai, Shen-Chuan; Huang, Jiun Jie; Chen, Peng Yu.

Proceedings - 2016 IEEE International Symposium on Computer, Consumer and Control, IS3C 2016. Institute of Electrical and Electronics Engineers Inc., 2016. p. 275-278 7545189 (Proceedings - 2016 IEEE International Symposium on Computer, Consumer and Control, IS3C 2016).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - A super-resolution algorithm using linear regression based on image self-similarity

AU - Tai, Shen-Chuan

AU - Huang, Jiun Jie

AU - Chen, Peng Yu

PY - 2016/8/16

Y1 - 2016/8/16

N2 - The application of image super-resolution technologies in recent years has increased noticeably. The main purpose of image up-scaling is to obtain high-resolution images from low-resolution images, and these up-scaled images should keep satisfactory visual qualities and present natural textures. The most popular image up-scaling algorithms are based on interpolation methods in spatial domain. However, the up-scaled images may produce blurring artifacts. Therefore, using spatial sharpening filters is usually used to make blurred images sharp and clear. The quantity of image sharpening is the key to decide the visual qualities of up-scaled images. In this paper, a method based on self-similarity of images and using simple linear regression to build a reconstruction model for improving visual qualities of up-scaled images adaptively is proposed. The experimental results show that our algorithm provides better subjective visual qualities as well as the peak signal-to-noise ratio (PSNR).

AB - The application of image super-resolution technologies in recent years has increased noticeably. The main purpose of image up-scaling is to obtain high-resolution images from low-resolution images, and these up-scaled images should keep satisfactory visual qualities and present natural textures. The most popular image up-scaling algorithms are based on interpolation methods in spatial domain. However, the up-scaled images may produce blurring artifacts. Therefore, using spatial sharpening filters is usually used to make blurred images sharp and clear. The quantity of image sharpening is the key to decide the visual qualities of up-scaled images. In this paper, a method based on self-similarity of images and using simple linear regression to build a reconstruction model for improving visual qualities of up-scaled images adaptively is proposed. The experimental results show that our algorithm provides better subjective visual qualities as well as the peak signal-to-noise ratio (PSNR).

UR - http://www.scopus.com/inward/record.url?scp=84986224228&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84986224228&partnerID=8YFLogxK

U2 - 10.1109/IS3C.2016.79

DO - 10.1109/IS3C.2016.79

M3 - Conference contribution

T3 - Proceedings - 2016 IEEE International Symposium on Computer, Consumer and Control, IS3C 2016

SP - 275

EP - 278

BT - Proceedings - 2016 IEEE International Symposium on Computer, Consumer and Control, IS3C 2016

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Tai S-C, Huang JJ, Chen PY. A super-resolution algorithm using linear regression based on image self-similarity. In Proceedings - 2016 IEEE International Symposium on Computer, Consumer and Control, IS3C 2016. Institute of Electrical and Electronics Engineers Inc. 2016. p. 275-278. 7545189. (Proceedings - 2016 IEEE International Symposium on Computer, Consumer and Control, IS3C 2016). https://doi.org/10.1109/IS3C.2016.79